Part 2: Teaching the Model About the Past Lag Features Calendar features: month, day of week, and is_weekend tell the model what time it is. That is useful for capturing seasonality and weekly rhythms, but it leaves out something equally important: what actually happened in the recent past. A store that sold 90 units last... Continue Reading →
Part 1: The Problem, the Data, and the Features For the past several posts, we built forecasting models the classical way: exponential smoothing, ARIMA, SARIMA. Each of those methods works on a single time series at a time, trying to model how a variable evolves through trend, seasonality, and noise. That framing is clean, but... Continue Reading →
The previous post established what stationarity means and how to recognise when a series lacks it. The airline passenger series fails on both counts: a clear upward trend and a repeating seasonal cycle, confirmed by an ACF that barely drops across 40 lags. This post is about what to do about it. The main tool... Continue Reading →
Toe Stop Drag Today I tried the toe stop drag, which is usually one of the first stopping methods people learn on quad skates. At first, the idea sounded simple: one foot stays in front carrying most of the weight, and the other foot drags behind using the toe stop. In practice, not that simple.... Continue Reading →
The previous posts built up a set of forecasting tools; SES, Holt-Winters, ARIMA, SARIMA, and each one quietly assumed something about the series it was working with. That assumption is stationarity. This post is about making that assumption explicit: what stationarity means, how to spot it, and what happens when a series violates it. Time... Continue Reading →
ARIMA captured the trend but left the seasonal cycle untouched. To handle both at once, SARIMA (Seasonal ARIMA) extends the model with a second set of parameters that operate at the seasonal lag rather than the observation lag. The result is a unified framework that handles trend and seasonality simultaneously. The parameters SARIMA(p,d,q)(P,D,Q)m has two... Continue Reading →
The previous post showed that AR, MA, and ARMA break down on the airline series because they assume stationarity. The common fix is to difference the series manually and then fit ARMA on the result. ARIMA formalises exactly this. The “I” stands for integrated, and it refers to the differencing step being built directly into... Continue Reading →
Acceleration (But Not Yet) Today I looked into acceleration. Not because I’m ready for it; I’m definitely not. Right now, I still don’t feel stable enough to go faster on purpose. So I’m continuing to focus on balance and control. But understanding acceleration early actually helps. It changes how you think about movement, even at... Continue Reading →
The smoothing methods from the previous posts; SES, DES, Holt-Winters work by taking weighted averages of past observations. This post introduces a different family: AR, MA, and ARMA models. Instead of smoothing, these treat forecasting as a regression problem. The predictors are either past observed values, past forecast errors, or a combination of both. The... Continue Reading →
SES models the level, while DES builds on this by incorporating a trend component. Naturally, this leads to the question of how to handle seasonality. For that reason, Triple Exponential Smoothing, better known as the Holt-Winters method—extends the DES framework by introducing a third component. Specifically, it adds a seasonal term that captures recurring patterns,... Continue Reading →
